• Gurpreet Kaur
• Dr. R.K. Bathla

One of the most common ways of image forgery is to embed a duplicate. In other words, this attack is called a copy-move. The embedding process consists of three stages: copying a fragment, adding changes to this fragment, and inserting a fragment into that area of the image whose contents are supposed to be hidden from the end user.In this paper, block based feature extraction and matching process is used. At first edge detection is carried out to get the high entropy pixels in the image so that matching process is carried out only for high entropy pixel blocks. Then feature extraction is carried out by converting image into overlapped blocks and mean and DCT features are extracted. Then mean values are put into a matrix and corresponding blocks are noted. Then mean value matrix is sorted in order to match the blocks with similar mean values. In matching process, variance of DCT features is used for similarity measure and forgery detection. Experimental results show that proposed method has high accuracy of forgery detection which comes in range of 97 to 99% along with least computation time.

Copy Move Forgery, Block Based Forgery Detection, Mean, Variance, DCT, LBP etc.

"DETECTING COPY MOVE FORGERY USING DCT AND MEAN VALUE OF BLOCKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.354-360, June 2019, Available :http://www.jetir.org/papers/JETIR1907I47.pdf

"DETECTING COPY MOVE FORGERY USING DCT AND MEAN VALUE OF BLOCKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp354-360, June 2019, Available at : http://www.jetir.org/papers/JETIR1907I47.pdf